Archives For

In an age of antitrust populism on both ends of the political spectrum, federal and state regulators face considerable pressure to deploy the antitrust laws against firms that have dominant market shares. Yet federal case law makes clear that merely winning the race for a market is an insufficient basis for antitrust liability. Rather, any plaintiff must show that the winner either secured or is maintaining its dominant position through practices that go beyond vigorous competition. Any other principle would inhibit the competitive process that the antitrust laws are designed to promote. Federal judges who enjoy life tenure are far more insulated from outside pressures and therefore more likely to demand evidence of anticompetitive practices as a predicate condition for any determination of antitrust liability.

This separation of powers between the executive branch, which prosecutes alleged infractions of the law, and the judicial branch, which polices the prosecutor, is the simple genius behind the divided system of government generally attributed to the eighteenth-century French thinker, Montesquieu. The practical wisdom of this fundamental principle of political design, which runs throughout the U.S. Constitution, can be observed in full force in the current antitrust landscape, in which the federal courts have acted as a bulwark against several contestable enforcement actions by antitrust regulators.

In three headline cases brought by the Department of Justice or the Federal Trade Commission since 2017, the prosecutorial bench has struck out in court. Under the exacting scrutiny of the judiciary, government litigators failed to present sufficient evidence that a dominant firm had engaged in practices that caused, or were likely to cause, significant anticompetitive effects. In each case, these enforcement actions, applauded by policymakers and commentators who tend to follow “big is bad” intuitions, foundered when assessed in light of judicial precedent, the factual record, and the economic principles embedded in modern antitrust law. An ongoing suit, filed by the FTC this year after more than 18 months since the closing of the targeted acquisition, exhibits similar factual and legal infirmities.

Strike 1: The AT&T/Time-Warner Transaction

In response to the announcement of AT&T’s $85.4 billion acquisition of Time Warner, the DOJ filed suit in 2017 to prevent the formation of a dominant provider in home-video distribution that would purportedly deny competitors access to “must-have” content. As I have observed previously, this theory of the case suffered from two fundamental difficulties. 

First, content is an abundant and renewable resource so it is hard to see how AT&T+TW could meaningfully foreclose competitors’ access to this necessary input. Even in the hypothetical case of potentially “must-have” content, it was unclear whether it would be economically rational for post-acquisition AT&T regularly to deny access to other distributors, given that doing so would imply an immediate and significant loss in licensing revenues without any clearly offsetting future gain in revenues from new subscribers.

Second, home-video distribution is a market lapsing rapidly into obsolescence as content monetization shifts from home-based viewing to a streaming environment in which consumers expect “anywhere, everywhere” access. The blockbuster acquisition was probably best understood as a necessary effort to adapt to this new environment (already populated by several major streaming platforms), rather than an otherwise puzzling strategy to spend billions to capture a market on the verge of commercial irrelevance. 

Strike 2: The Sabre/Farelogix Acquisition

In 2019, the DOJ filed suit to block the $360 million acquisition of Farelogix by Sabre, one of three leading airline booking platforms, on the ground that it would substantially lessen competition. The factual basis for this legal diagnosis was unclear. In 2018, Sabre earned approximately $3.9 billion in worldwide revenues, compared to $40 million for Farelogix. Given this drastic difference in market share, and the almost trivial share attributable to Farelogix, it is difficult to fathom how the DOJ could credibly assert that the acquisition “would extinguish a crucial constraint on Sabre’s market power.” 

To use a now much-discussed theory of antitrust liability, it might nonetheless be argued that Farelogix posed a “nascent” competitive threat to the Sabre platform. That is: while Farelogix is small today, it may become big enough tomorrow to pose a threat to Sabre’s market leadership. 

But that theory runs straight into a highly inconvenient fact. Farelogix was founded in 1998 and, during the ensuing two decades, had neither achieved broad adoption of its customized booking technology nor succeeded in offering airlines a viable pathway to bypass the three major intermediary platforms. The proposed acquisition therefore seems best understood as a mutually beneficial transaction in which a smaller (and not very nascent) firm elects to monetize its technology by embedding it in a leading platform that seeks to innovate by acquisition. Robust technology ecosystems do this all the time, efficiently exploiting the natural complementarities between a smaller firm’s “out of the box” innovation with the capital-intensive infrastructure of an incumbent. (Postscript: While the DOJ lost this case in federal court, Sabre elected in May 2020 not to close following similarly puzzling opposition by British competition regulators.) 

Strike 3: FTC v. Qualcomm

The divergence of theories of anticompetitive risk from market realities is vividly illustrated by the landmark suit filed by the FTC in 2017 against Qualcomm. 

The litigation pursued nothing less than a wholesale reengineering of the IP licensing relationships between innovators and implementers that underlie the global smartphone market. Those relationships principally consist of device-level licenses between IP innovators such as Qualcomm and device manufacturers and distributors such as Apple. This structure efficiently collects remuneration from the downstream segment of the supply chain for upstream firms that invest in pushing forward the technology frontier. The FTC thought otherwise and pursued a remedy that would have required Qualcomm to offer licenses to its direct competitors in the chip market and to rewrite its existing licenses with device producers and other intermediate users on a component, rather than device, level. 

Remarkably, these drastic forms of intervention into private-ordering arrangements rested on nothing more than what former FTC Commissioner Maureen Ohlhausen once appropriately called a “possibility theorem.” The FTC deployed a mostly theoretical argument that Qualcomm had extracted an “unreasonably high” royalty that had potentially discouraged innovation, impeded entry into the chip market, and inflated retail prices for consumers. Yet these claims run contrary to all available empirical evidence, which indicates that the mobile wireless device market has exhibited since its inception declining quality-adjusted prices, increasing output, robust entry into the production market, and continuous innovation. The mismatch between the government’s theory of market failure and the actual record of market success over more than two decades challenges the policy wisdom of disrupting hundreds of existing contractual arrangements between IP licensors and licensees in a thriving market. 

The FTC nonetheless secured from the district court a sweeping order that would have had precisely this disruptive effect, including imposing a “duty to deal” that would have required Qualcomm to license directly its competitors in the chip market. The Ninth Circuit stayed the order and, on August 11, 2020, issued an unqualified reversal, stating that the lower court had erroneously conflated “hypercompetitive” (good) with anticompetitive (bad) conduct and observing that “[t]hroughout its analysis, the district court conflated the desire to maximize profits with an intent to ‘destroy competition itself.’” In unusually direct language, the appellate court also observed (as even the FTC had acknowledged on appeal) that the district court’s ruling was incompatible with the Supreme Court’s ruling in Aspen Skiing Co. v. Aspen Highlands Skiing Corp., which strictly limits the circumstances in which a duty to deal can be imposed. In some cases, it appears that additional levels of judicial review are necessary to protect antitrust law against not only administrative but judicial overreach.

Axon v. FTC

For the most explicit illustration of the interface between Montesquieu’s principle of divided government and the risk posed to antitrust law by cases of prosecutorial excess, we can turn to an unusual and ongoing litigation, Axon v. FTC.

The HSR Act and Post-Consummation Merger Challenges

The HSR Act provides regulators with the opportunity to preemptively challenge acquisitions and related transactions on antitrust grounds prior to those transactions having been consummated. Since its enactment in 1976, this statutory innovation has laudably increased dealmakers’ ability to close transactions with a high level of certainty that regulators would not belatedly seek to “unscramble the egg.” While the HSR Act does not foreclose this contingency since regulatory failure to challenge a transaction only indicates current enforcement intentions, it is probably fair to say that M&A dealmakers generally assume that regulators would reverse course only in exceptional circumstances. In turn, the low prospect of after-the-fact regulatory intervention encourages the efficient use of M&A transactions for the purpose of shifting corporate assets to users that value those assets most highly.

The FTC’s Belated Attack on the Axon/Vievu Acquisition

Dealmakers may be revisiting that understanding in the wake of the FTC’s decision in January 2020 to challenge the acquisition of Vievu by Axon, each being a manufacturer of body-worn camera equipment and related data-management software for law enforcement agencies. The acquisition had closed in May 2018 but had not been reported through HSR since it fell well below the reportable deal threshold. Given a total transaction value of $7 million, the passage of more than 18 months since closing, and the insolvency or near-insolvency of the target company, it is far from obvious that the Axon acquisition posed a material competitive risk that merits unsettling expectations that regulators will typically not challenge a consummated transaction, especially in the case of what is a micro-sized nebula in the M&A universe. 

These concerns are heightened by the fact that the FTC suit relies on a debatably narrow definition of the relevant market (body-camera equipment and related “cloud-based” data management software for police departments in large metropolitan areas, rather than a market that encompassed more generally defined categories of body-worn camera equipment, law enforcement agencies, and data management services). Even within this circumscribed market, there are apparently several companies that offer related technologies and an even larger group that could plausibly enter in response to perceived profit opportunities. Despite this contestable legal position, Axon’s court filing states that the FTC offered to settle the suit on stiff terms: Axon must agree to divest itself of the Vievu assets and to license all of Axon’s pre-transaction intellectual property to the buyer of the Vievu assets. This effectively amounts to an opportunistic use of the antitrust merger laws to engage in post-transaction market reengineering, rather than merely blocking an acquisition to maintain the pre-transaction status quo.

Does the FTC Violate the Separation of Powers?

In a provocative strategy, Axon has gone on the offensive and filed suit in federal district court to challenge on constitutional grounds the long-standing internal administrative proceeding through which the FTC’s antitrust claims are initially adjudicated. Unlike the DOJ, the FTC’s first stop in the litigation process (absent settlement) is not a federal district court but an internal proceeding before an administrative law judge (“ALJ”), whose ruling can then be appealed to the Commission. Axon is effectively arguing that this administrative internalization of the judicial function violates the separation of powers principle as implemented in the U.S. Constitution. 

Writing on a clean slate, Axon’s claim is eminently reasonable. The fact that FTC-paid personnel sit on both sides of the internal adjudicative process as prosecutor (the FTC litigation team) and judge (the ALJ and the Commissioners) locates the executive and judicial functions in the hands of a single administrative entity. (To be clear, the Commission’s rulings are appealable to federal court, albeit at significant cost and delay.) In any event, a court presented with Axon’s claim—as of this writing, the Ninth Circuit (taking the case on appeal by Axon)—is not writing on a clean slate and is most likely reluctant to accept a claim that would trigger challenges to the legality of other similarly structured adjudicative processes at other agencies. Nonetheless, Axon’s argument does raise important concerns as to whether certain elements of the FTC’s adjudicative mechanism (as distinguished from the very existence of that mechanism) could be refined to mitigate the conflicts of interest that arise in its current form.

Conclusion

Antitrust vigilance certainly has its place, but it also has its limits. Given the aspirational language of the antitrust statutes and the largely unlimited structural remedies to which an antitrust litigation can lead, there is an inevitable risk of prosecutorial overreach that can betray the fundamental objective to protect consumer welfare. Applied to the antitrust context, the separation of powers principle mitigates this risk by subjecting enforcement actions to judicial examination, which is in turn disciplined by the constraints of appellate review and stare decisis. A rich body of federal case law implements this review function by anchoring antitrust in a decisionmaking framework that promotes the public’s interest in deterring business practices that endanger the competitive process behind a market-based economy. As illustrated by the recent string of failed antitrust suits, and the ongoing FTC litigation against Axon, that same decisionmaking framework can also protect the competitive process against regulatory practices that pose this same type of risk.

This guest post is by Jonathan M. Barnett, Torrey H. Webb Professor of Law at the University of Southern California, Gould School of Law.

State bar associations, with the backing of state judiciaries and legislatures, are typically entrusted with a largely unqualified monopoly over licensing in legal services markets. This poses an unavoidable policy tradeoff. Designating the bar as gatekeeper might protect consumers by ensuring a minimum level of service quality. Yet the gatekeeper is inherently exposed to influence by interests with an economic stake in the existing market. Any licensing requirement that might shield uninformed consumers from unqualified or opportunistic lawyers also necessarily raises an entry barrier that protects existing lawyers against more competition. A proper concern for consumer welfare therefore requires that the gatekeeper impose licensing requirements only when they ensure that the efficiency gains attributable to a minimum quality threshold outweigh the efficiency losses attributable to constraints on entry.

There is increasing reason for concern that state bar associations are falling short of this standard. In particular, under the banner of “legal ethics,” some state bar associations and courts have blocked or impeded entry by innovative “legaltech” services without a compelling consumer protection rationale.

The LegalMatch Case: A misunderstood platform

This trend is illustrated by a recent California appellate court decision interpreting state regulations pertaining to legal referral services. In Jackson v. LegalMatch, decided in late 2019, the court held that LegalMatch, a national online platform that matches lawyers and potential clients, constitutes an illegal referral service, even though it is not a “referral service” under the American Bar Association’s definition of the term, and the California legislature had previously declined to include online services within the statutory definition.

The court’s reasoning: the “marketing” fee paid by subscribing attorneys to participate in the platform purportedly runs afoul of state regulations that proscribe attorneys from paying a fee to referral services that have not been certified by the bar. (The lower court had felt differently, finding that LegalMatch was not a referral service for this purpose, in part because it did not “exercise any judgment” on clients’ legal issues.)

The court’s formalist interpretation of applicable law overlooks compelling policy arguments that strongly favor facilitating, rather than obstructing, legal matching services. In particular, the LegalMatch decision illustrates the anticompetitive outcomes that can ensue when courts and regulators blindly rely on an unqualified view of platforms as an inherent source of competitive harm.

Contrary to this presumption, legal services referral platforms enhance competition by reducing transaction-cost barriers to efficient lawyer-client relationships. These matching services benefit consumers that otherwise lack access to the full range of potential lawyers and smaller or newer law firms that do not have the marketing resources or brand capital to attract the full range of potential clients. Consistent with the well-established economics of platform markets, these services operate under a two-sided model in which the unpriced delivery of attorney information to potential clients is financed by the positively priced delivery of interested clients to subscribing attorneys. Without this two-sided fee structure, the business model collapses and the transaction-cost barriers to matching the credentials of tens of thousands of lawyers with the preferences of millions of potential clients are inefficiently restored. Some legal matching platforms also offer fixed-fee service plans that can potentially reduce legal representation costs relative to the conventional billable hour model that can saddle clients with unexpectedly or inappropriately high legal fees given the difficulty in forecasting the required quantity of legal services ex ante and measuring the quality of legal services ex post.

Blocking entry by these new business models is likely to adversely impact competition and, as observed in a 2018 report by an Illinois bar committee, to injure lower-income consumers in particular. The result is inefficient, regressive, and apparently protectionist.

Indeed, subsequent developments in this litigation are regrettably consistent with the last possibility. After the California bar prevailed in its legal interpretation of “referral service” at the appellate court, and the Supreme Court of California declined to review the decision, LegalMatch then sought to register as a certified lawyer referral service with the bar. The bar responded by moving to secure a temporary restraining order against the continuing operation of the platform. In May 2020, a lower state court judge both denied the petition and expressed disappointment in the bar’s handling of the litigation.

Bar associations’ puzzling campaign against “LegalTech” innovation

This case of regulatory overdrive is hardly unique to the LegalMatch case. Bar associations have repeatedly acted to impede entry by innovators that deploy digital technologies to enhance legal services, which can drive down prices in a field that is known for meager innovation and rigid pricing. Puzzlingly from a consumer welfare perspective, the bar associations have taken actions that impede or preclude entry by online services that expand opportunities for lawyers, increase the information available to consumers, and, in certain cases, place a cap on maximum legal fees.

In 2017, New Jersey Supreme Court legal ethics committees, following an “inquiry” by the state bar association, prohibited lawyers from partnering with referral services and legal services plans offered by Avvo, LegalZoom, and RocketLawyer. In 2018, Avvo discontinued operations due in part to opposition from multiple state bar associations (often backed up by state courts).

In some cases, bar associations have issued advisory opinions that, given the risk of disciplinary action, can have an in terrorem effect equivalent to an outright prohibition. In 2018, the Indiana Supreme Court Disciplinary Commission issued a “nonbinding advisory” opinion stating that attorneys who pay “marketing fees” to online legal referral services or agree to fixed-fee arrangements with such services “risk violation of several Indiana [legal] ethics rules.”

State bar associations similarly sought to block the entry of LegalZoom, an online provider of standardized legal forms that can be more cost-efficient for “cookie-cutter” legal situations than the traditional legal services model based on bespoke document preparation. These disputes are protracted and costly: it took LegalZoom seven years to reach a settlement with the North Carolina State Bar that allowed it to continue operating in the state. In a case pending before the Florida Supreme Court, the Florida bar is seeking to shut down a smartphone application that enables drivers to contest traffic tickets at a fixed fee, a niche in which the traditional legal services model is likely to be cost-inefficient given the relatively modest amounts that are typically involved.

State bar associations, with supporting action or inaction by state courts and legislatures, have ventured well beyond the consumer protection rationale that is the only potentially publicly-interested justification for the bar’s licensing monopoly. The results sometimes border on absurdity. In 2006, the New Jersey bar issued an opinion precluding attorneys from stating in advertisements that they had appeared in an annual “Super Lawyers” ranking maintained by an independent third-party publication. In 2008, based on a 304-page report prepared by a “special master,” the bar’s ethics committee vacated the opinion but merely recommended further consideration taking into account “legitimate commercial speech activities.” In 2012, the New York legislature even changed the “unlicensed practice of law” from a misdemeanor to a felony, an enhancement proposed by . . . the New York bar (see here and here). 

In defending their actions against online referral services, the bar associations argue that these steps are necessary to defend the public’s interest in receiving legal advice free from any possible conflict of interest. This is a presumptively weak argument. The associations’ licensing and other requirements are inherently tainted throughout by a “meta” conflict of interest. Hence it is the bar that rightfully bears the burden in demonstrating that any such requirement imposes no more than a reasonably necessary impediment to competition. This is especially so given that each bar association often operates its own referral service.

The unrealized potential of North Carolina State Board of Dental Examiners v. FTC

Bar associations might nonetheless take the legal position that they have statutory or regulatory discretion to take these actions and therefore any antitrust scrutiny is inapposite. If that argument ever held water, that is clearly no longer the case.

In an undeservedly underapplied decision, North Carolina State Board of Dental Examiners v. FTC, the Supreme Court held definitively in 2015 that any action by a “non-sovereign” licensing entity is subject to antitrust scrutiny unless that action is “actively supervised” by, and represents a “clearly articulated” policy of, the state. The Court emphasized that the degree of scrutiny is highest for licensing bodies administered by constituencies in the licensed market—precisely the circumstances that characterize state bar associations.

The North Carolina decision is hardly an outlier. It followed a string of earlier cases in which the Court had extended antitrust scrutiny to a variety of “hard” rules and “soft” guidance that bar associations had issued and defended on putatively publicly-interested grounds of consumer protection or legal ethics.

At the Court, the bar’s arguments did not meet with success. The Court rejected any special antitrust exemption for a state bar association’s “advisory” minimum fee schedule (Goldfarb v. Virginia State Bar (1975)) and, in subsequent cases, similarly held that limitations by professional associations on advertising by members—another requirement to “protect” consumers—do not enjoy any special antitrust exemption. The latter set of cases addressed specifically both advertising restrictions on price and quality by a California dental association (California Dental Association v. FTC (1999) ) and blanket restrictions on advertising by a bar association (Bates v. State Bar of Arizona (1977 )). As suggested by the bar associations’ recent actions toward online lawyer referral services, the Court’s consistent antitrust decisions in this area appear to have had relatively limited impact in disciplining potentially protectionist actions by professional associations and licensing bodies, at least in the legal services market. 

A neglected question: Is the regulation of legal services anticompetitive?

The current economic situation poses a unique historical opportunity for bar associations to act proactively by enlisting independent legal and economic experts to review each component of the current licensing infrastructure and assess whether it passes the policy tradeoff between protecting consumers and enhancing competition. If not, any such component should be modified or eliminated to elicit competition that can leverage digital technologies and managerial innovations—often by exploiting the efficiencies of multi-sided platform models—that have been deployed in other industries to reduce prices and transaction costs. These modifications would expand access to legal services consistent with the bar’s mission and, unlike existing interventions to achieve this objective through government subsidies, would do so with a cost to the taxpayer of exactly zero dollars.

This reexamination exercise is arguably demanded by the line of precedent anchored in the Goldfarb and Bates decisions in 1975 and 1977, respectively, and culminating in the North Carolina Dental decision in 2015. This line of case law is firmly grounded in antitrust law’s underlying commitment to promote consumer welfare by deterring collective action that unjustifiably constrains the free operation of competitive forces. In May 2020, the California bar took a constructive if tentative step in this direction by reviving consideration of a “regulatory sandbox” to facilitate experimental partnerships between lawyers and non-lawyers in pioneering new legal services models. This follows somewhat more decisive action by the Utah Supreme Court, which in 2019 approved commencing a staged process that may modify regulation of the legal services market, including lifting or relaxing restrictions on referral fees and partnerships between lawyers and non-lawyers.

Neither the legal profession generally nor the antitrust bar in particular has allocated substantial attention to potentially anticompetitive elements in the manner in which the practice of law has long been regulated. Restrictions on legal referral services are only one of several practices that deserve a closer look under the policy principles and legal framework set forth most recently in North Carolina Dental and previously in California Dental. A few examples can illustrate this proposition. 

Currently limitations on partnerships between lawyers and non-lawyers constrain the ability to achieve economies of scale and scope in the delivery of legal services and preclude firms from offering efficient bundles of complementary legal and non-legal services. Under a more surgical regulatory regime, legal services could be efficiently bundled with related accounting and consulting services, subject to appropriately targeted precautions against conflicts of interest. Additionally, as other commentators have observed and as “legaltech” innovations demonstrate, software could be more widely deployed to provide “direct-to-consumer” products that deliver legal services at a far lower cost than the traditional one-on-one lawyer-client model, subject to appropriately targeted precautions that reflect informational asymmetries in individual and small-business legal markets.

In another example, the blanket requirement of seven years of undergraduate and legal education raises entry costs that are not clearly justified for all areas of legal practice, some of which could potentially be competently handled by practitioners with intermediate categories of legal training. These are just two out of many possibilities that could be constructively explored under a more antitrust-sensitive approach that takes seriously the lessons of North Carolina Dental and the competitive risks inherent to lawyer self-regulation of legal services markets. (An alternative and complementary policy approach would be to move certain areas of legal services regulation out of the hands of the legal profession entirely.)

Conclusion

The LegalMatch case is indicative of a largely unexploited frontier in the application of antitrust law and principles to the practice of law itself. While commentators have called attention to the antitrust concerns raised by the current regulatory regime in legal services markets, and the evolution of federal case law has increasingly reflected these concerns, there has been little practical action by state bar associations, the state judiciary or state legislatures. This might explain why the delivery of legal services has changed relatively little during the same period in which other industries have been transformed by digital technologies, often with favorable effects for consumers in the form of increased convenience and lower costs. There is strong reason to believe a rigorous and objective examination of current licensing and related limitations imposed by bar associations in legal services markets is likely to find that many purportedly “ethical” requirements, at least when applied broadly and without qualification, do much to inhibit competition and little to protect consumers. 

This guest post is by Jonathan M. Barnett, Torrey H. Webb Professor Law, University of Southern California Gould School of Law.

It has become virtual received wisdom that antitrust law has been subdued by economic analysis into a state of chronic underenforcement. Following this line of thinking, many commentators applauded the Antitrust Division’s unsuccessful campaign to oppose the acquisition of Time-Warner by AT&T and some (unsuccessfully) urged the Division to take stronger action against the acquisition of most of Fox by Disney. The arguments in both cases followed a similar “big is bad” logic. Consolidating control of a large portfolio of creative properties (Fox plus Disney) or integrating content production and distribution capacities (Time-Warner plus AT&T) would exacerbate market concentration, leading to reduced competition and some combination of higher prices and reduced product for consumers. 

Less than 18 months after the closing of both transactions, those concerns seem to have been largely unwarranted. 

Far from precipitating any decline in product output or variety, both transactions have been followed by a vigorous burst of competition in the digital streaming market. In place of the Amazon plus Netflix bottleneck (with Hulu trailing behind), consumers now, or in 2020 will, have a choice of at least four new streaming services with original content, Disney+, AT&T’s “HBO Max”, Apple’s “Apple TV+” and Comcast’s NBCUniversal “Peacock” services. Critically, each service relies on a formidable combination of creative, financing and technological capacities that can only be delivered by a firm of sufficiently large size and scale.  As modern antitrust law has long recognized, it turns out that “big” is sometimes not bad.

Where’s the Harm?

At present, it is hard to see any net consumer harm arising from the concurrence of increased size and increased competition. 

On the supply side, this is just the next episode in the ongoing “Golden Age of Television” in which content producers have enjoyed access to exceptional funding to support high-value productions.  It has been reported that Apple TV+’s new “Morning Show” series will cost $15 million per episode while similar estimates are reported for hit shows such as HBO’s “Game of Thrones” and Netflix’s “The Crown.”  Each of those services is locked in a fierce competition to gain and retain sufficient subscribers to earn a return on those investments, which leads directly to the next happy development.

On the demand side, consumers enjoy a proliferating array of streaming services, ranging from free ad-supported services to subscription ad-free services. Consumers can now easily “cut the cord” and assemble a customized bundle of preferred content from multiple services, each of which is less costly than a traditional cable package and can generally be cancelled at any time.  Current market performance does not plausibly conform to the declining output, limited variety or increasing prices that are the telltale symptoms of a less than competitive market.

Real-World v. Theoretical Markets

The market’s favorable trajectory following these two controversial transactions should not be surprising. When scrutinized against the actual characteristics of real-world digital content markets, rather than stylized theoretical models or antiquated pre-digital content markets, the arguments leveled against these transactions never made much sense. There were two fundamental and related errors. 

Error #1: Content is Scarce

Advocates for antitrust intervention assumed that entry barriers into the content market were high, in which case it followed that the owner of an especially valuable creative portfolio could exert pricing power to consumers’ detriment. Yet, in reality, funding for content production is plentiful and even a service that has an especially popular show is unlikely to have sustained pricing power in the face of a continuous flow of high-value productions being released by formidable competitors. The amounts being spent on content in 2019 by leading streaming services are unprecedented, ranging from a reported $15 billion for Netflix to an estimated $6 billion for Amazon and Apple TV+ to an estimated $3.9 billion for AT&T’s HBO Max. It is also important to note that a hit show is often a mobile asset that a streaming or other video distribution service has licensed from independent production companies and other rights holders. Once the existing deal expires, those rights are available for purchase by the highest bidder. For example, in 2019, Netflix purchased the streaming rights to “Seinfeld”, Viacom purchased the cable rights to “Seinfeld”, and HBO Max purchased the streaming rights to “South Park.” Similarly, the producers behind a hit show are always free to take their talents to competitors once any existing agreement terminates.

Error #2: Home Pay-TV is a “Monopoly”

Advocates of antitrust action were looking at the wrong market—or more precisely, the market as it existed about a decade ago. The theory that AT&T’s acquisition of Time-Warner’s creative portfolio would translate into pricing power in the home pay-TV market mighthave been plausible when consumers had no reasonable alternative to the local cable provider. But this argument makes little sense today when consumers are fleeing bulky home pay-TV bundles for cheaper cord-cutting options that deliver more targeted content packages to a mobile device.  In 2019, a “home” pay-TV market is fast becoming an anachronism and hence a home pay-TV “monopoly” largely reduces to a formalism that, with the possible exception of certain live programming, is unlikely to translate into meaningful pricing power. 

Wait a Second! What About the HBO Blackout?

A skeptical reader might reasonably object that this mostly rosy account of the post-merger home video market is unpersuasive since it does not address the ongoing blackout of HBO (now an AT&T property) on the Dish satellite TV service. Post-merger commentary that remains skeptical of the AT&T/Time-Warner merger has focused on this dispute, arguing that it “proves” that the government was right since AT&T is purportedly leveraging its new ownership of HBO to disadvantage one of its competitors in the pay-TV market. This interpretation tends to miss the forest for the trees (or more precisely, a tree).  

The AT&T/Dish dispute over HBO is only one of over 200 “carriage” disputes resulting in blackouts that have occurred this year, which continues an upward trend since approximately 2011. Some of those include Dish’s dispute with Univision (settled in March 2019 after a nine-month blackout) and AT&T’s dispute (as pay-TV provider) with Nexstar (settled in August 2019 after a nearly two-month blackout). These disputes reflect the fact that the flood of subscriber defections from traditional pay-TV to mobile streaming has made it difficult for pay-TV providers to pass on the fees sought by content owners. As a result, some pay-TV providers adopt the negotiating tactic of choosing to drop certain content until the terms improve, just as AT&T, in its capacity as a pay-TV provider, dropped CBS for three weeks in July and August 2019 pending renegotiation of licensing terms. It is the outward shift in the boundaries of the economically relevant market (from home to home-plus-mobile video delivery), rather than market power concerns, that best accounts for periodic breakdowns in licensing negotiations.  This might even be viewed positively from an antitrust perspective since it suggests that the “over the top” market is putting pressure on the fees that content owners can extract from providers in the traditional pay-TV market.

Concluding Thoughts

It is common to argue today that antitrust law has become excessively concerned about “false positives”– that is, the possibility of blocking a transaction or enjoining a practice that would have benefited consumers. Pending future developments, this early post-mortem on the regulatory and judicial treatment of these two landmark media transactions suggests that there are sometimes good reasons to stay the hand of the court or regulator. This is especially the case when a generational market shift is in progress and any regulator’s or judge’s foresight is likely to be guesswork. Antitrust law’s “failure” to stop these transactions may turn out to have been a ringing success.

[TOTM: The following is the sixth in a series of posts by TOTM guests and authors on the FTC v. Qualcomm case recently decided by Judge Lucy Koh in the Northern District of California. Other posts in this series are here.

This post is authored by Jonathan M. Barnett, Torrey H. Webb Professor of Law at the University of Southern California Gould School of Law.]

There is little doubt that the decision in May 2019 by the Northern District of California in FTC v. Qualcomm is of historical importance. Unless reversed or modified on appeal, the decision would require that the lead innovator behind 3G and 4G smartphone technology renegotiate hundreds of existing licenses with device producers and offer new licenses to any interested chipmakers.

The court’s sweeping order caps off a global campaign by implementers to re-engineer the property-rights infrastructure of the wireless markets. Those efforts have deployed the instruments of antitrust and patent law to override existing licensing arrangements and thereby reduce the input costs borne by device producers in the downstream market. This has occurred both directly, in arguments made by those firms in antitrust and patent litigation or through the filing of amicus briefs, or indirectly by advocating that regulators bring antitrust actions against IP licensors.

Whether or not FTC v. Qualcomm is correctly decided largely depends on whether or not downstream firms’ interest in minimizing the costs of obtaining technology inputs from upstream R&D specialists aligns with the public interest in preserving dynamically efficient innovation markets. As I discuss below, there are three reasons to believe those interests are not aligned in this case. If so, the court’s order would simply engineer a wealth transfer from firms that have led innovation in wireless markets to producers that have borne few of the costs and risks involved in doing so. Members of the former group each exhibits R&D intensities (R&D expenditures as a percentage of sales) in the high teens to low twenties; the latter, approximately five percent. Of greater concern, the court’s upending of long-established licensing arrangements endangers business models that monetize R&D by licensing technology to a large pool of device producers (see Qualcomm), rather than earning returns through self-contained hardware and software ecosystems (see Apple). There is no apparent antitrust rationale for picking and choosing among these business models in innovation markets.

Reason #1: FRAND is a Two-Sided Deal

To fully appreciate the recent litigations involving the FTC and Apple on the one hand, and Qualcomm on the other hand, it is necessary to return to the origins of modern wireless markets.

Starting in the late 1980s, various firms were engaged in the launch of the GSM wireless network in Western Europe. At that time, each European telecom market typically consisted of a national monopoly carrier and a favored group of local equipment suppliers. The GSM project, which envisioned a trans-national wireless communications market, challenged this model. In particular, the national carrier and equipment monopolies were threatened by the fact that the GSM standard relied in part on patented technology held by an outside innovator—namely, Motorola. As I describe in a forthcoming publication, the “FRAND” (fair, reasonable and nondiscriminatory) principles that today govern the licensing of standard-essential patents in wireless markets emerged from a negotiation between, on the one hand, carriers and producers who sought a royalty cap and, on the other hand, a technology innovator that sought to preserve its licensing freedom going forward.

This negotiation history is important. Any informed discussion of the meaning of FRAND must recognize that this principle was adopted as something akin to a “good faith” contractual term designed to promote two objectives:

  1. Protect downstream adopters from holdup tactics by upstream innovators; and
  2. enable upstream innovators to enjoy an appreciable portion of the value generated by sales in the consumer market.

Any interpretation of FRAND that does not meet these conditions will induce upstream firms to reduce R&D investment, limit participation in standard-setting activities, or vertically integrate forward to capture directly a return on R&D dollars.

Reason #2: No Evidence of Actual Harm

In the December 2018 appellate court proceedings in which the Department of Justice unsuccessfully challenged the AT&T/Time-Warner merger, Judge David Sentelle of the D.C. Circuit said to the government’s legal counsel:

If you’re going to rely on an economic model, you have to rely on it with quantification. The bare theorem . . . doesn’t prove anything in a particular case.

The government could not credibly reply to that query in the AT&T case and, if appropriately challenged, could not do so in this case.

Far from being a market that calls out for federal antitrust intervention, the smartphone market offers what appears to be an almost textbook case of dynamic efficiency. For over a decade, implementers, along with sympathetic regulators and commentators, have argued that the market suffers (or, in a variation, will imminently suffer) from inflated prices, reduced output and delayed innovation as a result of “patent hold-up” and “royalty stacking” by opportunistic patent owners. In the course of several decades that have passed since the launch of the GSM network, none of these predictions have yet to materialize. To the contrary. The market has exhibited expanding output, declining prices (adjusted for increased functionality), constant innovation, and regular entry into the production market. Multiple empirical studies (e.g. this, this and this) have found that device producers bear on average an aggregate royalty burden in the single to mid-digits.

This hardly seems like a market in which producers and consumers are being “victimized” by what the Northern District of California calls “unreasonably high” licensing fees (compared to an unspecified, and inherently unspecifiable, dynamically efficient benchmark). Rather, it seems more likely that device producers—many of whom provided the testimony which the court referenced in concluding that royalty rates were “unreasonably high”—would simply prefer to pay an even lower fee to R&D input suppliers (with no assurance that any of the cost-savings would flow to consumers).

Reason #3: The “License as Tax” Fallacy

The rhetorical centerpiece of the FTC’s brief relied on an analogy between the patent license fees earned by Qualcomm in the downstream device market and the tax that everyone pays to the IRS. The court’s opinion wholeheartedly adopted this narrative, determining that Qualcomm imposes a tax (or, as Judge Koh terms it, a “surcharge”) on the smartphone market by demanding a fee from OEMs for use of its patent portfolio whether or not the OEM purchases chipsets from Qualcomm or another firm. The tax analogy is fundamentally incomplete, both in general and in this case in particular.

It is true that much of the economic literature applies monopoly taxation models to assess the deadweight losses attributed to patents. While this analogy facilitates analytical tractability, a “zero-sum” approach to patent licensing overlooks the value-creating “multiplier” effect that licensing generates in real-world markets. Specifically, broad-based downstream licensing by upstream patent owners—something to which SEP owners commit under FRAND principles—ensures that device makers can obtain the necessary technology inputs and, in doing so, facilitates entry by producers that do not have robust R&D capacities. All of that ultimately generates gains for consumers.

This “positive-sum” multiplier effect appears to be at work in the smartphone market. Far from acting as a tax, Qualcomm’s licensing policies appear to have promoted entry into the smartphone market, which has experienced fairly robust turnover in market leadership. While Apple and Samsung may currently dominate the U.S. market, they face intense competition globally from Chinese firms such as Huawei, Xiaomi and Oppo. That competitive threat is real. As of 2007, Nokia and Blackberry were the overwhelming market leaders and appeared to be indomitable. Yet neither can be found in the market today. That intense “gale of competition”, sustained by the fact that any downstream producer can access the required technology inputs upon payment of licensing fees to upstream innovators, challenges the view that Qualcomm’s licensing practices have somehow restrained market growth.

Concluding Thoughts: Antitrust Flashback

When competitive harms are so unclear (and competitive gains so evident), modern antitrust law sensibly prescribes forbearance. A famous “bad case” from antitrust history shows why.

In 1953, the Department of Justice won an antitrust suit against United Shoe Machinery Corporation, which had led innovation in shoe manufacturing equipment and subsequently dominated that market. United Shoe’s purportedly anti-competitive practices included a lease-only policy that incorporated training and repair services at no incremental charge. The court found this to be a coercive tie that preserved United Shoe’s dominant position, despite the absence of any evidence of competitive harm. Scholars have subsequently shown (e.g. this and  this; see also this) that the court did not adequately consider (at least) two efficiency explanations: (1) lease-only policies were widespread in the market because this facilitated access by smaller capital-constrained manufacturers, and (2) tying support services to equipment enabled United Shoe to avoid free-riding on its training services by other equipment suppliers. In retrospect, courts relied on a mere possibility theorem ultimately to order the break-up of a technological pioneer, with potentially adverse consequences for manufacturers that relied on its R&D efforts.

The court’s decision in FTC v. Qualcomm is a flashback to cases like United Shoe in which courts found liability and imposed dramatic remedies with little economic inquiry into competitive harm. It has become fashionable to assert that current antitrust law is too cautious in finding liability. Yet there is a sound reason why, outside price-fixing, courts generally insist that theories of antitrust liability include compelling evidence of competitive harm. Antitrust remedies are strong medicine and should be administered with caution. If courts and regulators do not zealously scrutinize the factual support for antitrust claims, then they are vulnerable to capture by private entities whose business objectives may depart from the public interest in competitive markets. While no antitrust fact-pattern is free from doubt, over two decades of market performance strongly favor the view that long-standing licensing arrangements in the smartphone market have resulted in substantial net welfare gains for consumers. If so, the prudent course of action is simply to leave the market alone.